htscoremain/htsalias: share the cmdl_* argv-block macros from htsalias.h

cmdl_room/cmdl_add/cmdl_ins were copy-pasted between htscoremain.c (the CLI
parser) and htsalias.c (config-file alias expansion), tagged "COPY OF cmdl_ins
in htscoremain.c". The copies had already drifted: htscoremain advanced the
pack offset by strlen+2, htsalias by strlen+1. Both are correct (a token plus
its NUL is L+1 bytes; +2 just leaves a one-byte gap), so the argv content was
identical either way, but two definitions of the same thing is one too many.

Move all three into htsalias.h (internal, gated by HTS_INTERNAL_BYTECODE,
already included by both translation units) and unify on the tight +1. This
only shrinks the inter-token gap in htscoremain's x_argvblk; every argv[] entry
is still an independently NUL-terminated string read through its own pointer,
so behavior is unchanged and the +32768 slack is untouched.

Adds 01_engine-doitlog.test for the doit.log reprise path, which drives
htscoremain's cmdl_ins (re-running httrack with no url re-inserts each recorded
argument) and had no coverage: 02_update-cache always passes a url, and
01_engine-rcfile exercises only the htsalias.c side. The test mirrors a file://
fixture, re-runs with no url, and asserts the reprise re-mirrors cleanly and
re-crawls the inserted url after a source change. Teeth-checked: dropping the
+1 makes the inserted tokens run together and the test fails on the resulting
crawl error.

make check: 16 PASS, 7 SKIP (offline). shellcheck clean.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Signed-off-by: Xavier Roche <roche@httrack.com>

src/htsalias.c

@@ -41,25 +41,6 @@ Please visit our Website: http://www.httrack.com
 
 #define _NOT_NULL(a) ( (a!=NULL) ? (a) : "" )
 
-// COPY OF cmdl_ins in htscoremain.c
-/* Bytes left in x_argvblk from offset ptr. The offset can in principle outrun
-   the block (alias/doit.log expansion), so the copy aborts cleanly instead of
-   the subtraction wrapping to a huge unbounded size. */
-#define cmdl_room(bufsize, ptr)                                                \
-  ((ptr) < (size_t) (bufsize) ? (size_t) (bufsize) - (ptr) : 0)
-// Insert a command in the argc/argv (buff has total capacity bufsize)
-#define cmdl_ins(token, argc, argv, buff, bufsize, ptr)                        \
-  {                                                                            \
-    int i;                                                                     \
-    for (i = argc; i > 0; i--)                                                 \
-      argv[i] = argv[i - 1];                                                   \
-  }                                                                            \
-  argv[0] = (buff + ptr);                                                      \
-  strlcpybuff(argv[0], token, cmdl_room(bufsize, ptr));                        \
-  ptr += (int) (strlen(argv[0]) + 1);                                          \
-  argc++
-// END OF COPY OF cmdl_ins in htscoremain.c
-
 /*
   Aliases for command-line and config file definitions
   These definitions can be used:

src/htsalias.h

@@ -52,6 +52,34 @@ const char *opttype_value(int p);
 const char *opthelp_value(int p);
 const char *hts_gethome(void);
 void expand_home(String * str);
+
+/* Command-line argv-block builders, shared by htscoremain.c (the CLI parser)
+   and htsalias.c (config-file alias expansion). Tokens are packed back-to-back
+   into x_argvblk (total capacity bufsize); each argv[] entry points into the
+   block. cmdl_room bounds every copy: the running offset ptr can outrun the
+   block (alias / doit.log expansion outpacing the +32768 slack), so it yields
+   0 rather than a wrapped size_t and the bounded copy aborts cleanly. */
+#define cmdl_room(bufsize, ptr)                                                \
+  ((ptr) < (size_t) (bufsize) ? (size_t) (bufsize) - (ptr) : 0)
+
+/* Append a token as a new argv[argc]. */
+#define cmdl_add(token, argc, argv, buff, bufsize, ptr)                        \
+  argv[argc] = (buff + ptr);                                                   \
+  strlcpybuff(argv[argc], token, cmdl_room(bufsize, ptr));                     \
+  ptr += (int) (strlen(argv[argc]) + 1);                                       \
+  argc++
+
+/* Insert a token at argv[0], shifting the existing argc entries up by one. */
+#define cmdl_ins(token, argc, argv, buff, bufsize, ptr)                        \
+  {                                                                            \
+    int i;                                                                     \
+    for (i = argc; i > 0; i--)                                                 \
+      argv[i] = argv[i - 1];                                                   \
+  }                                                                            \
+  argv[0] = (buff + ptr);                                                      \
+  strlcpybuff(argv[0], token, cmdl_room(bufsize, ptr));                        \
+  ptr += (int) (strlen(argv[0]) + 1);                                          \
+  argc++
 #endif
 
 #endif

src/htscoremain.c

@@ -69,31 +69,6 @@ Please visit our Website: http://www.httrack.com
 /* Resolver */
 extern int IPV6_resolver;
 
-/* Remaining room in the argv block; 0 once it is exhausted (alias expansion or
-   doit.log insertion can outrun the +32768 slack), so the copy aborts cleanly
-   instead of the subtraction wrapping to a huge unbounded size. */
-#define cmdl_room(bufsize, ptr)                                                \
-  ((ptr) < (size_t) (bufsize) ? (size_t) (bufsize) - (ptr) : 0)
-
-// Add a command in the argc/argv (buff has total capacity bufsize)
-#define cmdl_add(token, argc, argv, buff, bufsize, ptr)                        \
-  argv[argc] = (buff + ptr);                                                   \
-  strlcpybuff(argv[argc], token, cmdl_room(bufsize, ptr));                     \
-  ptr += (int) (strlen(argv[argc]) + 2);                                       \
-  argc++
-
-// Insert a command in the argc/argv (buff has total capacity bufsize)
-#define cmdl_ins(token, argc, argv, buff, bufsize, ptr)                        \
-  {                                                                            \
-    int i;                                                                     \
-    for (i = argc; i > 0; i--)                                                 \
-      argv[i] = argv[i - 1];                                                   \
-  }                                                                            \
-  argv[0] = (buff + ptr);                                                      \
-  strlcpybuff(argv[0], token, cmdl_room(bufsize, ptr));                        \
-  ptr += (int) (strlen(argv[0]) + 2);                                          \
-  argc++
-
 #define htsmain_free() do { \
   if (url != NULL) { \
     free(url); \

tests/01_engine-doitlog.test

@@ -0,0 +1,92 @@
+#!/bin/bash
+#
+
+# doit.log reprise (no network). Re-running httrack in a mirror directory with
+# NO url makes the engine read hts-cache/doit.log and re-insert every recorded
+# argument into the command line through htscoremain.c's cmdl_ins macro (the
+# x_argvblk builder). That path is distinct from the rc-file one in
+# 01_engine-rcfile.test (htsalias.c) and from the url-on-command-line update in
+# 02_update-cache.test, and nothing else exercises it. Two properties:
+#   1. A multi-token reprise re-mirrors cleanly: every token (the url and each
+#      option) survives the back-to-back packing, so the no-url run reproduces
+#      the file set with no errors. A packing/bound bug corrupts a later token
+#      and surfaces as an error or a missing file.
+#   2. The reprise actually re-crawls through the inserted url: changing a source
+#      file and re-running with no url picks up the new content.
+
+set -euo pipefail
+
+# Resolve httrack to an absolute path before we cd: PATH may hold a
+# build-relative entry that would not resolve from the temp directory.
+bin=$(command -v httrack) || {
+    echo "FAIL: httrack not found on PATH"
+    exit 1
+}
+case "$bin" in
+/*) ;;
+*) bin="$(cd "$(dirname "$bin")" && pwd)/$(basename "$bin")" ;;
+esac
+
+tmp=$(mktemp -d "${TMPDIR:-/tmp}/httrack_doitlog.XXXXXX") || exit 1
+trap 'rm -rf "$tmp"' EXIT HUP INT QUIT PIPE TERM
+
+site="$tmp/site"
+out="$tmp/out"
+mkdir -p "$site/sub"
+printf '<a href="a.html">a</a> <a href="sub/b.html">b</a>' >"$site/index.html"
+echo 'OLDCONTENT' >"$site/a.html"
+echo '<p>bbb</p>' >"$site/sub/b.html"
+url="file://$site/index.html"
+
+# count Error: lines in the log (grep -c exits 1 on zero matches: guard it)
+errors() { grep -ciE '^[0-9:]*[[:space:]]Error:' "$out/hts-log.txt" || true; }
+
+# initial mirror with the url and a handful of options, so doit.log records a
+# multi-token command line for cmdl_ins to re-insert one token at a time.
+rc=0
+"$bin" "$url" -O "$out" --quiet -n -%v0 -r3 >/dev/null 2>&1 || rc=$?
+test "$rc" -eq 0 || {
+    echo "FAIL: initial mirror exited $rc"
+    exit 1
+}
+test -f "$out/hts-cache/doit.log" || {
+    echo "FAIL: doit.log not written by the initial mirror"
+    exit 1
+}
+
+# --- 1. no-url reprise re-mirrors cleanly -----------------------------------
+# No url on the command line, so the engine loads doit.log and re-inserts the
+# recorded arguments (cmdl_ins). -O selects the mirror; argv carries no url.
+rc=0
+"$bin" -O "$out" --quiet >/dev/null 2>&1 || rc=$?
+test "$rc" -eq 0 || {
+    echo "FAIL: doit.log reprise exited $rc"
+    exit 1
+}
+test "$(errors)" = 0 || {
+    echo "FAIL: doit.log reprise reported errors (a token may have been corrupted)"
+    grep -iE 'Error:' "$out/hts-log.txt" | head -3
+    exit 1
+}
+for suffix in a.html sub/b.html; do
+    test -n "$(find "$out" -path "*/$suffix" -print -quit)" || {
+        echo "FAIL: $suffix missing after the no-url reprise"
+        exit 1
+    }
+done
+
+# --- 2. the reprise re-crawls through the inserted url -----------------------
+sleep 1
+echo 'NEWCONTENT' >"$site/a.html"
+rc=0
+"$bin" -O "$out" --quiet >/dev/null 2>&1 || rc=$?
+test "$rc" -eq 0 || {
+    echo "FAIL: second reprise exited $rc"
+    exit 1
+}
+grep -q NEWCONTENT "$(find "$out" -path '*/a.html' -print -quit)" || {
+    echo "FAIL: reprise did not pick up the changed source (inserted url not re-crawled)"
+    exit 1
+}
+
+exit 0

tests/Makefile.am

@@ -19,6 +19,7 @@ TESTS = \
 	01_engine-cache.test \
 	01_engine-charset.test \
 	01_engine-cmdline.test \
+	01_engine-doitlog.test \
 	01_engine-entities.test \
 	01_engine-filter.test \
 	01_engine-hashtable.test \